Technical Unsolvable cyl #5 misfire - Tried EVERYTHING

wheela

Lieutenant
Jun 4, 2021
1,082
599
0
Twin Cities, MN
Ride
2015 e84 X1 35i Msport
Need to read the manual in this case, to get exactly how the engine registers a misfire. I know what a misfire is, but I assumed that knock sensors played a role in their detection by somehow figuring out that a cylinder "sounded" different to the others so it misfired but probably I am wrong. Thanks for the clarification, will research this further.
This is basically how it detects knock/pre-ignition, as this makes a "ping" sound/vibration that the knock sensors pick up. But for misfire, the DME watches crank rotation sensor (or some sensor that reads rotating speed of the engine) and it can see the small acceleration from each cylinder's power stroke as the engine rotates. So if there is a misfire, it can see that the crank didn't accelerate as expected during a particular cylinder's power stroke.
 

martymil

Major General
Sep 6, 2017
3,331
1,906
0
Down Under
Ride
S65 1m
The stock DCT flywheels are a big problem at 20 psi+, anything post that you need a s55 one.

Seen this happen on many cars here in oz with DCT.
 

Jern54

Sergeant
Oct 18, 2019
287
1
166
0
Netherlands
Ride
E91 N54 335i
I'm using a Adamat flywheel since I have a 8 bolt crank.
You can choose between multiple spring ratings.
I've went with 1050nm rated springs.

Clutch chatter is barely noticeable and it works really well at 700whp.
 

Catalin335xi

Lurker
Jun 29, 2023
20
6
0
Romania
Ride
E92 335xi N54, built engine
Sad update: I know in the post description I said that I don't have blowby on the injector, well I checked today and apparently I do have blowby on injector 5, which causes compression loss during WOT on high rpm and boost. So mistery solved... need a new cylinder head.

Tried using teflon tape along with new decoupler and gasket and the issue persists. Pic with the naughty injector:

20230720_202645.jpg
 

IQraceworks

Corporal
Jul 7, 2020
203
111
0
Missouri
Ride
07' BMW 335i
Can't you just do the inject bore fix with the threaded bushings to fix the injector bore?

Any maybe this is a dumb question...but did you see if a new injector tip seal would fix the blowby issue?
 

Catalin335xi

Lurker
Jun 29, 2023
20
6
0
Romania
Ride
E92 335xi N54, built engine
Is there any other kit other than the one from ppifpr.com? that guy is a scam, I've already ordered something from him since january and received nothing, no refund and on top of it he stopped replying, I am going through the bank chargeback now. If there are other places or methods for this I am all ears but I do not know another method
 
  • Informative
Reactions: wheela

zaw111

Specialist
Oct 2, 2019
67
13
0
St-Petersburg
Ride
Zxcvbnm1
I have the same problem with 6 cyl . 26 psi ...I tried everything too With lover boost OK ... Turbosystems turbos . I had totai engine boooom and replace engine - the same ....it sometimes appears when accelerating WOT in 4th gear at 7000... I wrote to Ken Wedge and he also said the injector seat. Perhaps it is 6 that breaks because it is the most loaded. I bought another block head from disassembly. They cost us only 150 dollars. Let's see what happens
 

135boost

Corporal
Oct 28, 2017
242
1
133
0
Finland
Ride
135is
Long time without anything to write caused of a friend blowing my engine.. about this misfire, i think it's origin is same than what i han some years ago.. nothing to do with mechanical parts, just 1 value hidden in software.. i'm sitting in a car so i can't dig it out now, just wrote this cause if that misfire is still on the table i have a way to find this topic
 

135boost

Corporal
Oct 28, 2017
242
1
133
0
Finland
Ride
135is
Long time without anything to write caused of a friend blowing my engine.. about this misfire, i think it's origin is same than what i han some years ago.. nothing to do with mechanical parts, just 1 value hidden in software.. i'm sitting in a car so i can't dig it out now, just wrote this cause if that misfire is still on the table i have a way to find this topic
So, if misfire happend at prox 6000rpm, wot, boost levels over levels 27psi and engine tuned obvius rich, check both bank afr values.

If one of those jump to 16+ at least with e85, problem can be on dme logic and it's not cyl 5, it's whole bank runnin extreme lean and for some logic reason cyl 5 trigs misfire easier than those other 2.

You don't find that map on any public xdf and you can't monitor it with any public logger so tracking down it is close to impossible to track down root cause of it.

I did some 500hr logs and 547maps with awful much mechanical tests to solve the issue before with good friends figured out logic behind this
 

135boost

Corporal
Oct 28, 2017
242
1
133
0
Finland
Ride
135is
This map is where i think you find root cause for mysterious misfires.

For testing purpose only and to understand what you manipulate, this is air mass table to trig calculated limits to send dme to path that cause erratic misfires.

It is to be found from dme logic description page 16xx, 41xx and 68xx..
 

Attachments

  • 1699941246269569153186648529571.jpg
    1699941246269569153186648529571.jpg
    518.5 KB · Views: 49

wheela

Lieutenant
Jun 4, 2021
1,082
599
0
Twin Cities, MN
Ride
2015 e84 X1 35i Msport
Long time without anything to write caused of a friend blowing my engine.. about this misfire, i think it's origin is same than what i han some years ago.. nothing to do with mechanical parts, just 1 value hidden in software.. i'm sitting in a car so i can't dig it out now, just wrote this cause if that misfire is still on the table i have a way to find this topic
Hi 135boost. In your perusing through the dme, have you come across any tables that define the window/period for misfire detection regarding what rotational position the dme is looking to see each pistons power stroke? I'm curious if the false misfires at high power levels could be related to increasing crank twist as power increases? Just a theory I have - here's what I wrote in a different thread about it. If this is happening, it seems you could adjust the window to compensate with the right tables.
 

Attachments

  • Screenshot_20231114_091100_Chrome.jpg
    Screenshot_20231114_091100_Chrome.jpg
    197.7 KB · Views: 41

135boost

Corporal
Oct 28, 2017
242
1
133
0
Finland
Ride
135is
on my tests, one key factor what i trigged misfires was that air mass value on right corner of that table.
i think orginal dme keep values over those 400 implausable and that trig some safety measures.

my engine had air mass value on prox 400 when it every time trigged misfire and if allowed recovery
without engine stop start, it waked up with a 0.8 sec misfire/1,3 sec recovery/0,8sec misfire sequense.
if i loveres boost, it trigged that at 6200rpm, if full boost, it came at 5500rpm ( remember now that i havefull boost at 5000)

what i think it did was "resetting" some amount maps to base value like at least fuel scalars.
i monitored injector windows on all phases, there at those rpm levels it run with single phase injection
so it was easy to log injection time to drop on whole bank from prox 250 to value 170.
what make it difficult to figure out the cause was that on a bank there is 3 cylinders so if one is shut off
cause of misfire, it deduce 1/3 of bank fuel quantity what is seen on AFR as 1/3 jump upward from
11.5 to some 17 levels.

that reduction from 250 to 170 is exact same size drop so i tought at the beginning that it`s 1 cylinder
issue but in fact it`s whole bank was running extreme lean.
17:1 mixture it selves cause misfires and knock and i`m not sure about logic why it seems to be cyulinder 5 and 2
what seems to be most prone to trig misfire indication.

my best guess is that if dme jump on some "emergy mode" caused on implausable air mass reading, it run on
reduced map reading on other maps than just fuel scalar, like knock cylinder individual separating.
there was then at least 3 things what this emergency mode demand to trig
-wot
-rich mixture
-400 air mass

if that happend and knock logic start to read bank separated , closest cylinders to knock sensors are ... drumroll... 2 and 5..
and when 1 of those 3 demand was not fullified and normal mode was accepted without stop start, engine recover from
that state.. untill all demands again was fullified.

that state what dissapeared was "is engine running rich"

i got huge amount of help from here anbd did some over 500hr real log testing with modifications
to track that issue down and my heureka moment came when i got on mhd logger possibility to log
injector timing versus knock logic activating.

at the end, tested to mod that table of air mass at rpm over 5000 and row 160 with increase to over 1000.
then first ever test pull, 3rd to 4th gear, all erratic ignition pulls away, first pull to redline at 8100rpm, transmission clutch axle broke
so sure, there was a difference in pull.

i will combine here picture from my phone, a logic "map" what i did and i think there happend, sorry of misspellings etc, take a look and comment if my logic sucks..

PS.
i`m not sure of the rest of you guys, but there was on someone issue with injector bore wear out, only thing that
can cause wear is movement, so if that bore is wearing, injector have to vibrate in a bore.
i did calculations on cylinder pressure changes and peak pressures, measured that injector spring and came with
end conclusion that i was not happy about spring load/pressure pushing injector upwards safety margin so
i use on my engine helicoil thread in cylinder head and double springs stacked on those injectors
 
  • Like
Reactions: wheela

wheela

Lieutenant
Jun 4, 2021
1,082
599
0
Twin Cities, MN
Ride
2015 e84 X1 35i Msport
Wow, so increasing the air mass values in that table seems to have fixed it? It looks like x-axis is rpm, do you know what y-axis is?

Your cylinder pressure calculations sound really interesting. Did you some kind of back-calculation from torque, or find some first principals equations? I'm curious at what approximate point (load%? torque level? Etc.) do you estimate cylinder pressure can start to move the injectors against the spring?
 

135boost

Corporal
Oct 28, 2017
242
1
133
0
Finland
Ride
135is
I used cylinder pressure variation calculator and calculated injector area, then measured spring pressure.
Conclution, spring was suitable for maybe 2,5 times of stock cylinder pressure.

About that map, did some estimations about y axis but at the end, i dit test and error on that one.